databricks-jobs
Pass
Audited by Gen Agent Trust Hub on Aug 11, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [SAFE]: The skill documents official Databricks tools and libraries, providing standard templates for job orchestration and monitoring.- [PROMPT_INJECTION]: The skill contains examples for processing external parameters that represent a surface for indirect prompt injection, specifically SQL injection.
- Ingestion points: The example
process_region.pyinexamples.mdingests data viadbutils.widgets.get("region"). - Boundary markers: No validation or boundary markers are demonstrated in the example code.
- Capability inventory: Ingested parameters are directly used in
spark.sql()execution. - Sanitization: The code uses f-string interpolation for SQL queries without sanitization, which is an insecure pattern for untrusted input.
Audit Metadata